iv.] VARIOUS THEORIES, &c. 99 



so vigorous as to drain all the liquid water from the 

 cavities, the current may possibly pass as imbibed 

 water. In those cases where Dufour could not press 

 water through his sharply bent shoots, the lumina of 

 the vessels were not quite closed, however nearly they 

 might have been ; all that occurred was a curtailing of 

 the supply of water for the time being, whence 

 continued transpiration rarefied the air above the bend 

 to such an extent that enough water was squeezed 

 through to keep the shoot from drooping. In those 

 cases where Dufour could only drive very little water 

 beyond the sharp bend, it was because the cells, with 

 their rarefied air bubbles, retained most of the water 

 absorbed it and held it fast. 



As to Dufour's criticism that the sum of the air- 

 pressures in the tree must be less than an atmosphere, 

 and therefore cannot lift water more than 10 meters 

 high, that must be conceded ; but the pressure of the 

 air-bubbles does not perform the lifting of the columns, 

 its work is rather to distend the elastic closing mem- 

 branes of the bordered pits, so as to enable water to 

 filter through them. The lifting is performed chiefly 

 by capillarity, which raises the column of water 

 particles in each tracheide. 



One difficulty avowedly arises (but the imbibition 

 theory does not explain it) that is, why is it, when 



H 2 



